2022-05579 - Fault localization in wind turbine blades based on model updating
Level of qualifications required : Graduate degree or equivalent
Fonction : Internship Research
AssignmentDescription
A solution to increase the lifespan and reduce the maintenance cost of wind turbines is to go towards efficient predictive maintenance. It requires the detection of faults (cracks, ice on a blade etc), their localisation and the quantification of their size to engage or not repair actions.
When a fault appears, the structural properties of the wind turbine are impacted most of the time (natural frequencies, mode shapes etc). Using only sensor measurements and data-driven approaches, fault detection can be done by tracking the change in structural features that can be extracted from the monitored signals. Thus, when a large change is observed, a fault is assumed and an alarm can be triggered. The next step is to localise the fault, exploiting a model of the structure (a finite element model e.g.). This is transformed into an optimisation problem, where the model is parametrised and then fitted to match the measurements (the process of model updating ). It is assumed that the parametrisation can model the damage. The size of the problem makes the optimisation problem computationally expensive and difficult to solve. To reduce this cost, many methods have investigated evolutionary algorithms, such as CMA-ES or genetic algorithms.
This assumes that the model is exact and represents reality. However, data are noisy due to environmental variation and the model is not perfect due to manufacturing tolerances or wear for example, and because the range of physical phenomena considered is necessarily limited. This uncertainty must be considered in the model updating to fit the model under noisy and uncertain data assumptions. In this context, Bayesian model updating techniques allow for this consideration. The objective of the internship is to investigate the potential that offers these methods for model updating of wind turbines, that experience a strongly noisy environment.
The intern will have access to two models, an academic model of a beam with a reduced numerical cost and a finite element model of a wind turbine blade to generate simulated data, and to in-house algorithms for the identification of structural features from sensor measurements. He/she will assess the efficiency of Bayesian updating methods on the academic model first for different settings (e.g. position and number of sensors, operating conditions, tracked features etc.). In the second step, the method will be extended to the full 3D FEM of the wind turbine blade.
Objectives
Supervision context
The internship will take place at the Inria Centre of Rennes.
This research internship is part of a collaborative project between Inria (French Institut national de recherche en sciences et technologies du numérique ) and IFPEN ( Institut Français du Pétrole et des Energies Nouvelles ) on wind turbine monitoring. The trainee will therefore have the opportunity to interact with researchers and engineers from both institutes. The supervision team will be composed of E. Denimal (Inria), L. Mevel (Inria), J-L Pfister (IFPEN) and M.R. El Amri (IFPEN).
There is a possibility to pursue this internship with a PhD.
Main activitiesObjectives
Skills and profile
Theme/Domain : Optimization, machine learning and statistical methods Scientific computing (BAP E)
Town/city : Rennes
Inria is the French national research institute dedicated to digital science and technology. It employs 2,600 people. Its 200 agile project teams, generally run jointly with academic partners, include more than 3,500 scientists and engineers working to meet the challenges of digital technology, often at the interface with other disciplines. The Institute also employs numerous talents in over forty different professions. 900 research support staff contribute to the preparation and development of scientific and entrepreneurial projects that have a worldwide impact.
Instruction to applyMerci de déposer en ligne CV, lettre de motivation et éventuelles recommandations
Pour plus d'information, contactez [email protected]
Defence Security : This position is likely to be situated in a restricted area (ZRR), as defined in Decree No. 2011-1425 relating to the protection of national scientific and technical potential (PPST).Authorisation to enter an area is granted by the director of the unit, following a favourable Ministerial decision, as defined in the decree of 3 July 2012 relating to the PPST. An unfavourable Ministerial decision in respect of a position situated in a ZRR would result in the cancellation of the appointment.
Recruitment Policy : As part of its diversity policy, all Inria positions are accessible to people with disabilities.
Warning : you must enter your e-mail address in order to save your application to Inria. Applications must be submitted online on the Inria website. Processing of applications sent from other channels is not guaranteed.